IBM-AIX: MIGRATION ISSUES

Showing posts with label MIGRATION ISSUES. Show all posts

Wednesday, April 8, 2020

Migration issues - After the OS migration done, we could not able to ssh to the server?

We could not able to ssh to the server once we migrated the server to AIX 7.1. Looks like there is a bug in AIX 7.1 ssh.

Please follow the below to fix the issue.

Problem:

=======

testlpar: # ssh root@testlpar2

Connection closed by testlpar2

testlpar: #

Solution:

=======

Take the backup of /etc/ssh/sshd_config file and please add "hostKeyAlgorithms +ssh-dss" to the /etc/ssh/sshd_config file, then you can able to ssh to the server witout any issues.

Take the backup of /etc/ssh/sshd_config file and please add "HostKey /etc/ssh/ssh_host_rsa_key" to the /etc/ssh/sshd_config file, then you can able to ssh to the server witout any issues.

Now try to ssh now!

testlpar: # ssh root@testlpar2

root@testlpar2's password:

Migration issues - Server degraded but lppchk didn't show anything in aix?

After the server migration done, the OS level is degraded to 7.1.3.4. when i checked the "lppchk -vm3" it did not show any output.

so I tried to check with oslevel -rl command and identified the issue. Here is the history and solution

testlpar: # oslevel -s

7100-03-04-1441

testlpar: #

testlpar: # lppchk -vm3 -> lppchk didn't show any error. looks like clean

testlpar: #

testlpar: # oslevel -rl 7100-05 -> This will clearly tell you where the issue is.

Fileset Actual level Recommended level

Java5.sdk 5.0.0.570 5.0.0.620

testlpar: #

testlpar: # instfix -i|grep -i ml

Not all filesets for 7100-04_AIX_ML were found

Not all filesets for 7100-05_AIX_ML were found

testlpar: #

Solution:

It clearly says that the problem is with the java5.sdk filesets. Speak to IBM and get the Java5.sdk packages as it is not comes in default when you download the packages from fixcentral for the migration.

Keep the packages under /tmp/java_package directory

cd /tmp/java_packages

inutoc .

installp -acgXYd . java5.sdk

lppchk -vm3

oslevel -rl 7100-05 -> Now the oslevel will be in the expected desired level

instfix -i|egrep -i 'ml|sp' -> All the ML and SP filesets were found.

Migration issues - lppchk issues after migration done in aix?

testlpar: # lppchk -vm3

lppchk: The following filesets need to be installed or corrected to bring the system to a consistent state.

bos.ecc_client.rte 7.1.5.0 (usr: APPLIED, root: BROKEN)

rpm.rte 4.13.0.2 (usr: COMMITED, root: not installed)

testlpar: #

Solution:

=======

testlpar: # cd <directory of AIX 7.1.5.2 lppsource directory which you used for migratioin>

you may see the below files under it.

testlpar: # ls -al bos.ecc*

-rwxr-xr-x 1 root system 2356343 APR 11 2020 bos.ecc_client.7.1.5.0.I

-rwxr-xr-x 1 root system 2356343 APR 11 2020 bos.ecc_client.7.1.5.0.U

testlpar: # ls -al rpm*

-rwxr-xr-x 1 root system 4832243 APR 11 2020 rpm.rte.4.13.0.2.I

testlpar: #

Then force re-isntall them as below.

testlpar: # installp -acFXYd ./ bos.ecc_client.rte

testlpar: # installp -acgNXYd ./ bos.ecc_client.rte 7.1.5.0

testlpar: # installp -acFXYd ./ rpm.rte

Then run "lppchk -vm3", it should be clean.

Miration issues: After migration done, we could not able to ssh to the server due to the changes on the /.ssh/known_hosts in aix?

Miration issues: After migration is done, we could not able to ssh to the server due to the changes in the /.ssh/known_hosts in aix?

After the server migrated, we could not ablet o ssh the server due to the changes on the /.ssh/known_hosts during the migration, this situation would happen especially on the cluster environment. In order to fix this, you have to follow the error message as the solution is mentioned in the error messages clearly.

testlpar1: # ssh root@testlpar2

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!

someone could be eavesdropping on you right now (man-in-the-middle attack)!

It is also possible that a host key has just been changed.

The fingerprint for the RSA Key sent by the remote host is

SHA256:Qx5ofuasgihsknglisidghi4gosidgnog/LQg.

Please correct your system administrator.

Add correct host key in /.ssh/known_hosts to get rid of this message.

Offending DSA key in /.ssh/known_hosts:3

RSA host key for tesetlpar2 has changed and you have requested strict checking.

Host key verification failed.

testlpar1: #

Solution

=======

In the above errors, it clearly says that we need to correct the host key in /.ssh/known_hosts and it also provide the solution like "offending DSA key in /.ssh/known_hosts:3" which means

that you need to comment out the third (3) lines on the /.ssh/known_hosts using vi editor.

cp -p /.ssh/known_hosts /.ssh/known_hosts_backup

vi /.ssh/known_hosts (comment out the 3rd line)

The first time it asks the yes/no confirmation, next time onwards it won't ask. (directly logon to the servers)

Tuesday, April 7, 2020

Migration issues - "Not all the filesets for 71-02-031334_SP were found" in aix?

After the migration done, we got the below error when we execute the "instfix -i|grep -i sp" command.

"Not all the filesets for 71-02-031334_SP were found"

Solution:

In order to fix this, first we need to identify what filesets are not found in the particular SP level. We can use the below command to identify the filesets details of the particular sp level.

instfix -icqk <not found filesets level> |grep ":-:"

testlpar: # instfix -icqk 71-02-031334SP|grep ":-:"

71-02-031334_SP: DirectorCommonAgent:6.3.0.3:6.2.1.3:-:AIX 7100-02-03 service pack

testlpar: #

so the above output clearly says that the service pack issue with the "DirectorCommonAgent" which is not required, so we can removed it to fix it.

Migration issues - After reboot, The server went to open firmware state in aix?

After the server reboot, the server went to open firmware state.

When I checked one the console, it look like below.

"testlpar, open firmware, 2"

Solution

========

Power off the server and then power on the server with teh normal mode. (The issue occured as the server is set with "Open the server in SMS mode")

lssyscfg -r lpar -F name,state,lpar_id

chsysstate -r lapr -o shutdown --id 2

chsysstate -r lpar -o on --id 2 -b norm

mkvt -id 2

(or)

chsysstate -m <frame_name> -r lpar -o shutdown --immed -n <lpar_name>

chsysstate -r lpar -m <frame_name> -o on -f Normal -b norm -n <lpar_name>

Migration issues - niminit command failed in aix?

testlpar: # niminit -a master=testnim -a name=testlpar

0042-037 NIMkid: The state of "Hostname and Service name not provided or found" prevents this operation from succeeding.

testlpar: #

Solution:

=======

Logon to the nim server and confirmed that the lpar entry is on the NIM's /etc/hosts file, If not there, then add it., and then define the machine on the nim server, It will fix the issue.

Migration issues - Broken director common agent failure on lppchk output in aix?

We are planning to migrate from 7.1.4.5 to 7.1.5.2. When i checked "lppchk -vm3", I can see the below errors.

testlpar: # lppchk -vm3

lppchk: The following filesets need to be installed or corrected to bring the system to a consistent state.

DirectorcommonAgent 6.3.5.0 (usr: COMMITTED, root:BROKEN)

testlpar: #

Solution:

========

We can remove the direcotrcommonagent filesets as it is not required.

testlpar: # installp -u DirectorCommonAgent

testlpar: # installp -ug DirectorCommonAgent

Migration issues - rsh issue on the nimadmvg in aix?

testnim: # nimadm -j nimadmvg -c testlpar -s aix710502_spot -l aix710502_lppsource -d hdisk1 –Y

Initializing the NIM master

Initializing the NIM client testlpar

rshd: 0826-826 The host name for your address is not known

0505-159 nimadm: warning,unexpected result from remote command to testlpar.

0505-153 nimadm: unable to execute remote client commands.

Cleaning up alt_disk_migration on the NIM master.

Done!!!

testnim: #

Troubleshoot

============

testnim: # rsh testlpar uname -a

rshd: 0826-826 The host name for your address is not known.

testnim: #

Solution

========

Check the nim servers details (hostname-IP address) are updated on the client's /etc/hosts and vice versa.

Remove the /etc/niminfo file and recreate with niminit command (If required disable the and enable the rsh again) and then try again.

migration issue: After upgrade to AIX 7.1.5.2 - the oslevel showing as downgraded that is 7100-04-06-1806 and lppchk display rpm.rte filesets in broken state.

testlpar: # lppchk -v

lppchk: the following filesets need to be installed or corrected to bring the system to a consistent state.

rpm.rte 3.0.5.52 (usr: COMMITED, root:BROKEN)

testlpar: #

testlpar: # lslpp -l|grep -i rpm.rte

rpm.rte 3.0.5.52 COMMITTED RPM package manager

rpm.rte 3.0.5.52 BROKEN RPM Package manager

testlpar: #

Solution:

=======

Mount the 7.1.5.2 lppsource and perform the smitty update_all. It will fix the rpm.rte issue and we will reach to 7.1.5.2